8‐Hydroxyquinolines are bactericidal against Mycobacterium tuberculosis

ABSTRACT There is an urgent need for new treatments effective against Mycobacterium tuberculosis, the causative agent of tuberculosis. The 8‐hydroxyquinoline series is a privileged scaffold with anticancer, antifungal, and antibacterial activities. We conducted a structure–activity relationship study of the series regarding its antitubercular activity using 26 analogs. The 8‐hydroxyquinolines showed good activity against M. tuberculosis, with minimum inhibitory concentrations (MIC90) of <5 μM for some analogs. Small substitutions at C5 resulted in the most potent activity. Substitutions at C2 generally decreased potency, although a sub‐family of 2‐styryl‐substituted analogs retained activity. Representative compounds demonstrated bactericidal activity against replicating M. tuberculosis with >4 log kill at 10× MIC over 14 days. The majority of the compounds demonstrated cytotoxicity (IC50 of <100 μM). Further development of this series as antitubercular agents should address the cytotoxicity liability. However, the 8‐hydroxyquinoline series represents a useful tool for chemical genomics to identify novel targets in M. tuberculosis.


| INTRODUCTION
Tuberculosis is the leading killer among infectious diseases with 1.3 million deaths and >10 million new cases in 2017 (World Health Organization, 2018). The shortest treatment for tuberculosis is a 6-month regimen containing four antibiotics (World Health Organization, 2018).
More than half a million multidrug resistant (MDR) cases emerge each year, with treatment of those cases requiring up to 2 years of treatment with second and third line drugs (World Health Organization, 2018).
Affordable and innovative treatments with shorter treatment times and unique mechanisms of action are urgently needed.

| Cytotoxicity
HepG2 human liver cells (ATCC) were seeded in 384-well plates at 1,800 cells per well and incubated in a humidified atmosphere of 37 C, 5% CO 2 in DMEM (Invitrogen), 10% FBS, 1 mM sodium pyruvate, 2 mM Glutagro (Corning), 25 mM glucose, 100 I.U/mL penicillin, and 100 μg/mL streptomycin. Compounds were added 24 hr post cell seeding to 1% DMSO final concentration. After a 72 hr incubation period, CellTiter-Glo ® Reagent was added to the 384-well plates. Relative luminescent units (RLUs) were measured using a Biotek Synergy 4 plate reader. Raw data were normalized by the average RLU value from 1% DMSO treated wells and expressed as percent growth.
Growth inhibition curves were fitted using the Levenberg-Marquardt algorithm. The IC 50 was defined as the compound concentration that produced 50% of the growth inhibitory response.

| Determination of minimum inhibitory concentration
M. tuberculosis strains were grown in Middlebrook 7H9 supplemented with 10% v/v oleic acid, albumin, dextrose, and catalase (OADC) (Becton Dickinson) and 0.05% w/v Tween 80 or on Middlebrook 7H10 agar supplemented with 10% v/v OADC. MICs were determined as previously described (Ollinger et al., 2013). In brief, compounds were tested for activity at varying concentrations against a

| Kill kinetics
Compounds 24 and 4 were tested for their ability to kill M. tuberculosis in replicating conditions (Early & Alling, 2015). Compounds were added to M. tuberculosis cultures in liquid medium and viability monitored over 3 weeks standing at 37 C by plating for colony-forming units (CFU).
Plates were incubated at 37 C for 4 weeks before colonies were counted.

| SAR studies
The HQ series has good activity against M. tuberculosis in vitro. We, and others, identified these compounds in phenotypic screening as hits (Ananthan et al., 2009;Ollinger et al., 2018). We wanted to explore the SAR for this series to determine the scope for further development. We designed a set of compounds representing three general chemotypes built around the HQ scaffold: (a) substituted at the 5-position (Table 1), (b) substituted at the 2-position (Table 2), and (c) styryl substituents at the 2-position (Table 3). For each compound, we determined activity against M. tuberculosis by measuring the minimum inhibitory concentration (MIC 90 ), defined as the concentration required to inhibit growth by 90%. We also determined cytotoxicity against the HepG2 cell line by measuring the IC 50 , defined as the concentration required to reduce cell viability by 50%. The seed compound, unsubstituted 8-hydroxyquinoline (1), had an MIC of 3.6 μM and HepG2 IC 50 of 7.6 μM.
Our exploration began with the introduction of substituents at the 5-position (Table 1). Compounds with methyl (2), bromide (3), chloride (4), and nitro (6)  Finally, the scaffold proved broadly tolerant to substitution with styryl groups at the 2-position, albeit with the induction of significant variation between replicates (Table 3). Substitutions on the styryl ring (20-25) led to only minor changes in potency. As with the active compounds from the previous two chemotypes, these compounds were consistently cytotoxic. Replacement of the vinyl substructure with a piperazine (26) was not tolerated.

| Kill kinetics
We determined the kill kinetics of two active compounds that were randomly selected against aerobically-grown, actively replicating (for 24 at 6.8 μM and for 4 at 5.8 μM). It is likely that this represents outgrowth of resistant mutants at the later time point. Alternatively, it could reflect compound instability over the extended exposure period, but that has not been tested.

| Cytotoxicity
The main liability for this series is cytotoxicity. Three analogs (14,15,18) showed no measurable cytotoxicity against HepG2 cells, but none of these were active against M. tuberculosis. All 19 of the active compounds were cytotoxic for this cell line. To compare cell lines, all 26 analogs were tested for cytotoxicity against Vero cells; all active analogs were cytotoxic in this cell line as well (Table 4). These cytotoxicity data are in contrast to results from other groups, where cytotoxicity for 8-hydroxyquinoline (1) and related compounds was not noted (Ananthan et al., 2009;Darby & Nathan, 2010;Shah et al., 2016). The differences in these data may reflect the status of cells in the cytotoxicity assay, since, similar to our findings, Shen et al. found HQ analogs to be cytotoxic via inhibition of DNA synthesis in HepG2 cells (Shen, Chen, & Roffler, 1999). Since our assay uses replicating HepG2 or Vero cells to measure toxicity, we would expect to see negative effects from the inhibition of DNA synthesis and by extension cell division. We were unable to identify analogs that lacked toxicity against replicating HepG2 or Vero cells. Further work to identify the mechanism of action of the HQ compounds, which may or may not involve metal chelation, might reveal a way forward in designing analogs with specificity for mycobacteria (Prachayasittikul et al., 2013;Shen et al., 1999).

| CONCLUSION
In conclusion, we confirmed that the HQ series have good antitubercular properties in vitro with rapid bactericidal activity. SAR demonstrated that small substitutions at the 5-position were tolerated, but substitution at the 2-position was not generally tolerated, except for 2-styrenyl-substitutions. Further work to elucidate the bacterial target could pave a way forward in designing analogs without cytotoxicity.

ACKNOWLEDGMENTS
This study was funded in part by Eli Lilly and Company in support of the mission of the Lilly TB Drug Discovery Initiative and with funding from the Bill and Melinda Gates Foundation under grant OPP1024038.
The authors thank the following for technical assistance: James Ahn, Torey Alling, Devon Dennison, Jack Elder, Lindsay Flint, Megha Gupta, Stephanie Florio, Douglas Joerss, Steven Mullen, Catie Shelton, Dean Thompson, and James Vela.

CONFLICT OF INTEREST
The authors declare they have no conflict of interest.